Drug-induced QT-interval prolongation and resultant sudden cardiac death due to ventricular arrhythmia is an uncommon but devastating side effect of more than 70 currently marketed drugs, including multiple commonly-used antipsychotic and antidepressant medications. Most recently, the US FDA issued a warning that the most widely prescribed antidepressant in the US, citalopram, has been associated with QT prolongation, an effect confirmed in recent investigations using electronic health records. Prevention of drug-induced arrhythmia has focused on identifying at-risk drugs. However, there are also characteristics of the vulnerable patient, including common genetic variation, that place certain individuals at particularly high risk of QT prolongation and fatal arrhythmia. Identification of these at-risk individuals represents an important gap in current clinical knowledge. Electrocardiographic QT interval is heritable and has a graded relationship to sudden cardiac death and to arrhythmias from medications. In particular, 16 common genetic variants in twelve genes have been demonstrated to influence inter-individual variability in QT duration; individuals in the top and bottom quintiles of a score of 14 variants have an approximately 10-15 msec difference in QT interval, equivalent to the degree of QT prolongation of some non- cardiac medications withdrawn from the market for arrhythmias. To better estimate the clinical impact of the common variants associated with QT duration, the present study will investigate these variants in 3 complementary contexts. First, building on previous work by this group, electronic health records (EHR) will be used to identify patients with unusually long or short QT intervals; discarded blood will be genotyped for common QT variants. Second, 80 healthy volunteers will be identified on the basis of a QT genotype score to receive moxifloxacin, a marketed antibiotic that prolongs the QT interval. Finally, EHR will be used to identify individuals who were exposed to psychotropic treatments, had ECGs pre and post, and prospective collection of discarded blood samples will be genotyped for known QT-prolonging variants. These studies will allow more precise estimation of the risk associated with common genetic variation and their potential additive/synergistic effects of QT-prolonging medications. Existing medications as well as future ones may be used more safely if this risk can be quantified.